The British Biophysical Society

EPR- Electron Paramagnetic Resonance

Chris Cooper

 

E = Electrons

These are the useful bits of atoms. Most of physics, all of chemistry and therefore all of biochemistry, biology, behaviour, life etc. involves the transfer of electrons. Electrons are small, negatively charged, and determine the chemical reactivity of atoms and molecules.

P = Paramagnetic

Electrons also possess the property of spin or paramagnetism. Thus the bulk property we view as magnetism, for example, is a consequence of the properties of the 'spin' on individual electrons. Almost all electrons go around in pairs, but some are solitary. These 'unpaired electrons' are found in free radicals (highly reactive molecules) and transition metals ions (e.g. iron, copper, cobalt).

R = Resonance

When things vibrate at the same frequency they are in resonance - or 'in tune' - and small signals can be picked out (e.g. tuning a radio) and enlarged (e.g. pushing a swing).

An EPR experiment

So when we do an EPR experiment we are looking at the few unpaired electrons that are present in the sample. We put the sample in a box in a variable magnetic field. We send radio waves into the box (cavity) down a pipe (waveguide). The source of the radio (microwave) signal is tuned so that the waves fit into the box exactly. They bounce backwards and forwards in the cavity before they escape back up the waveguide to the detector. The magnetic part of the radio wave excites the unpaired electrons most when they are in their favourite magnetic field. They then soak up some of the microwaves so less get back to the detector. There is a characteristic magnetic field where this microwave is absorbed. This value when combined with the frequency of the microwaves fed into the cavity and some physical constants is used to work out the 'g value' for the unpaired electrons.

An isolated electron has a g value of almost exactly 2 (2.002319304386 if you're a trainspotter). This ideal value is never found in a real molecule as the spin of an unpaired electron 'senses' other electrons and nucleii in its nearby environment and this changes the observed g-value. So far this is physics and therefore utterly reliable but boring (uninformative). Where does the chemistry (moderately interesting and moderately comprehensible) and the biology and medicine (very interesting but very incomprehensible) come in?

Biomedical EPR spectroscopy

As the spin of an unpaired electron can sense it's nearby environment each free radical or transition metal ion will have slightly different properties. Therefore EPR can be used to identify biological molecules that contain free radicals or transition metal ions in their structure. Even more usefully, EPR is a quantitative technique i.e. we can determine the concentration of unpaired electrons present in a sample even if we do not know the exact nature of the free radical being observed. This is very important in, for example, distinguishing between reactive free radicals that are present in high concentrations and may be damaging and those that may be present in only very low concentrations and may not be. EPR spectroscopy has been shown to be useful in understanding the pathophysiology and underlying chemical mechanism of a wide range of diseases. Examples studied at the University of Essex include Parkinson's Disease, birth asphyxia, stroke, septic shock, kidney damage and coronary heart disease.

The two major types of EPR signals of biomedical relevance are unpaired electrons in free radicals and metalloproteins. We can illustrate the usefulness of EPR with reference to signals found in blood.

Metalloproteins

Many blood proteins use metal centres to bind and react with small molecules. For example haemoglobin contains iron and gives red blood cells their colour. The iron group in haemoglobin transports oxygen around the body. One form of haemoglobin is not able to bind oxygen (ferric or methaemoglobin) and can carry out dangerous side-reactions with small biological molecules called peroxides. EPR can detect the concentration of methaemoglobin more accurately than any other technique and we see elevated amounts in patients with septic shock in intensive care units. Metmyoglobin, a related protein found in muscle cells, is also present in excess concentrations in the kidney following muscle trauma, a condition termed rhabdomyolysis, and is responsible for the resulting renal failure.

We can also detect EPR signals from the protein transferrin that transports iron around the body and a copper-containing protein called ceruloplasmin. Catalase, an enzyme that degrades the toxic peroxides mentioned above, is also detectable using the new high-sensitivity EPR instrumentation currently available. The relationship of these signals to different disease states is currently under active investigation. A typical EPR spectrum of blood is illustrated below.

Free Radicals

Free radicals are reactive molecules that are implicated in a number of disease processes. In the blood we have recently identified the dominant free radical as being bound to the molecule haemoglobin described above. This 'globin radical' is formed when methaemoglobin reacts with peroxides and can damage the membranes that keep biological cells intact (see below). This damage can lead to atherosclerotic lesions and ultimately to coronary heart disease.

The free radical in blood is identical to that seen when peroxide is added to haemoglobin (adapted from Svistunenko et al., 1997a and 1997b)

Not all free radicals are bad. Nitric oxide (NO) is a free radical that is used as a normal physiological messenger molecule by the body. Unfortunately there are some pathophysiological conditions where the NO concentration rises to toxic levels. One such example is when a new-born baby has suffered a drop in blood flow to the brain. Unfortunately NO is not readily detectable by EPR as it has a very broad signal. However, it can be made detectable by binding it to a biological molecule that has a more distinct and long-lived EPR signal. This is the basis of the commonly-used EPR spin-trapping technique. In the case of NO nature has its own intrinsic spin trap, haemoglobin. NO displaces oxygen from haemoglobin and generates a readily detectable haem-nitrosyl signal. Alternatively, exogenously added iron complexes (such as DETC) are available to trap NO as a ferrous nitrosyl complex. Studies on these signals are contributing to our knowledge of the etiology of a number of disease processes and assist in monitoring therapeutic strategies.

Spectrum from DETC - a nitric oxide spin trap

Reading List

Svistunenko, D.A., Davies, N.A., Wilson, M.T., Stidwill, R.P., Singer, M. and Cooper, C.E. (1997a). Free radical in blood: a measure of haemoglobin autoxidation in vivo? J. Chem. Soc. Perkin Trans. II 2, 2539-2543.

Svistunenko, D.A., Patel, R.P., Voloshchenko, S.V. and Wilson, M.T. (1997b). The globin-based free radical of ferryl hemoglobin is detected in normal human blood. J. Biol. Chem. 272, 7114-7121.

Chris Cooper is Professor of Biochemistry at the University of Essex. He is also the director of the University of Essex Biomedical EPR facility. If you would like to contact Chris, send an email to ccooper@essex.ac.uk.

Chris would like to thank Barnard for assistance with the text, Dimitri Svistunenko for help with the figures and the Wellcome Trust for funding.